Recent advances in molecular biology have provided the opportunity to identify pathogens, diagnose disease states, and perform forensic determinations using gene sequences specific for the desired purpose. This explosion of genetic information has created a need for high-capacity assays and equipment for performing molecular biological assays, particularly nucleic acid hybridization assays. Most urgently, there is a need to miniaturize, automate, standardize and simplify such assays. This need stems from the fact that while these hybridization assays were originally developed in research laboratories working with purified products and performed by highly skilled individuals, adapting these procedures to clinical uses, such as diagnostics, forensics and other applications, has produced the need for equipment and methods that allow less-skilled operators to effectively perform the assays under higher capacity, less stringent assay conditions.
Existing technology utilizes the binding of molecules contained within a biologically reactive sample fluid, hereinafter referred to as target molecules, onto molecules contained within biologically reactive sites, hereinafter referred to as probe molecules. The primary enabler of this technology is an apparatus commonly referred to as a biochip, which comprises one or more ordered microscopic arrays (“microarrays”) of biologically reactive sites immobilized on the surface of a substrate. A biologically reactive site can be created by dispensing a small volume of a fluid containing a biological reagent onto a discrete location on the surface of a substrate, also commonly referred to as spotting. To enhance immobilization of probe molecules, biochips can include a 2-dimensional array of 3-dimensional polymeric anchoring structures (for example, polyacrylamide gel pads) attached to the surface of the substrate. Probe molecules such as oligonucleotides are covalently attached to polyacrylamide-anchoring structures by forming amide, ester or disulfide bonds between the biomolecule and a derivatized polymer comprising the cognate chemical group. Covalent attachment of probe molecules to such polymeric anchoring structures is usually performed after polymerization and chemical cross-linking of the polymer to the substrate is completed.
Of particular interest are methods of analyzing the nucleic acid in a sample of cells. The conventional way of analyzing the nucleic acid present in a sample of cells involves performing multiple steps using several different bench top instruments in a laboratory setting. First, the nucleic acid must be extracted from the cells in the sample. This is typically done by performing any number of cell lysing procedures that cause the cells to break apart and release their contents. Next, the nucleic acid is typically separated from the rest of the cell contents, as the presence of other cell contents may be undesirable in subsequent steps. Frequently, a nucleic acid amplification reaction is done to obtain suitable amounts of nucleic acid for characterization. The resulting amplified nucleic acid products can then be identified by any number of techniques.
There are a variety of nucleic acid amplification reactions that are used, some of which utilize thermal cycling. Briefly, these techniques can be classified as either target amplification or signal amplification. Target amplification involves the amplification (i.e. replication) of the target sequence to be detected, resulting in a significant increase in the number of target molecules. Target amplification strategies include the polymerase chain reaction (PCR), strand displacement amplification (SDA), nucleic acid sequence based amplification (NASBA), and transcription mediated amplification (TMA).
Alternatively, rather than amplify the target, alternate techniques use the target as a template to replicate a signalling probe, allowing a small number of target molecules to result in a large number of signalling probes, that then can be detected. Signal amplification strategies include the ligase chain reaction (LCR), cycling probe technology (CPT), Invader™, Q-beta replicase (QBR), and the use of “amplification probes” such as “branched DNA” that result in multiple label probes binding to a single target sequence.
The polymerase chain reaction (PCR) is widely used and described, and involve the use of primer extension combined with thermal cycling to amplify a target sequence. This technique has been applied to a wide variety of biological methods, including for example, DNA sequence analysis, probe generation, cloning of nucleic acid sequences, directed mutagenesis, detection of genetic mutations, diagnoses of viral infections, molecular “fingerprinting,” and the monitoring of contaminating microorganisms. See U.S. Pat. Nos. 4,683,195 and 4,683,202, and PCR Essential Data, J. W. Wiley & sons, Ed. C. R. Newton, 1995, all of which are incorporated by reference. In addition, there are a number of variations of PCR which may also find use in the invention, including “quantitative competitive PCR” or “QC-PCR”, “arbitrarily primed PCR” or “AP-PCR”, “immuno-PCR”, “Alu-PCR”, “PCR single strand conformational polymorphism” or “PCR-SSCP”, “reverse transcriptase PCR” or “RT-PCR”, “biotin capture PCR”, “vectorette PCR”, “panhandle PCR”, and “PCR select cDNA subtration”, among others.
The polymerase chain reaction comprises repeated rounds, or cycles, of target denaturation, primer annealing, and extension. This reaction process yields an exponential amplification of the desired target sequence and is most advantageously accomplished through the use of a thermally-stable polymerase. The length of time required to complete a particular PCR protocol is dependent upon the number of amplification cycles as well as the length of the denaturation, annealing, and extension steps. A typical PCR performed on a conventional thermal cycler can often take several hours.
The fidelity and efficiency of PCR amplification is affected by several factors. These factors include the concentration of various reaction components, particularly the polymerase, deoxynucleotide triphosphates, magnesium ions, target molecules, and amplimers (amplification primer pair), the length and temperature of the denaturation, annealing, and extension steps, the number of cycles, and the specificity and length of the amplimers. Since the success of any given PCR amplification depends upon a number of variables, optimized reaction conditions are often empirically determined. However, such an optimization process is usually labor intensive, costly, and time consuming.
Strand displacement amplification (SDA) is generally described in Walker et al., in Molecular Methods for Virus Detection, Academic Press, Inc., 1995, and U.S. Pat. Nos. 5,455,166 and 5,130,238, all of which are hereby incorporated by reference.
Nucleic acid sequence based amplification (NASBA) is generally described in U.S. Pat. No. 5,409,818; Sooknanan et al., Nucleic Acid Sequence-Based Amplification, Ch. 12 (pp. 261-285) of Molecular Methods for Virus Detection, Academic Press, 1995; and “Profiting from Gene-based Diagnostics”, CTB International Publishing Inc., N.J., 1996, both of which are incorporated by reference.
Transcription mediated amplification (TMA) is generally described in U.S. Pat. Nos. 5,399,491, 5,888,779, 5,705,365, 5,710,029, all of which are incorporated by reference.
Cycling probe technology (CPT) is a nucleic acid detection system based on signal or probe amplification rather than target amplification, such as is done in polymerase chain reactions (PCR). Cycling probe technology relies on a molar excess of labeled probe which contains a scissile linkage of RNA. Upon hybridization of the probe to the target, the resulting hybrid contains a portion of RNA:DNA. This area of RNA:DNA duplex is recognized by RNAseH and the RNA is excised, resulting in cleavage of the probe. The probe now consists of two smaller sequences which may be released, thus leaving the target intact for repeated rounds of the reaction. The unreacted probe is removed and the label is then detected. CPT is generally described in U.S. Pat. Nos. 5,011,769, 5,403,711, 5,660,988, and 4,876,187, and PCT published applications WO 95/05480, WO 95/1416, and WO 95/00667, all of which are specifically incorporated herein by reference.
The ligation chain reaction (LCR) involve the ligation of two smaller probes into a single long probe, using the target sequence as the template for the ligase. See generally U.S. Pat. Nos. 5,185,243 and 5,573,907; EP 0 320 308 B1; EP 0 336 731 B1; EP 0 439 182 B1; WO 90/01069; WO 89/12696; and WO 89/09835, all of which are incorporated by reference.
Q-beta replicase (QBR) is a mRNA amplification technique, similar to NASBA and TMA, that relies on an RNA-dependent RNA polymerase derived from the bacteriophage Q-beta that can synthesize up to a billion stands of product from a template.
Invader™ technology is based on structure-specific polymerases that cleave nucleic acids in a site-specific manner. Two probes are used: an “invader” probe and a “signalling” probe, that adjacently hybridize to a target sequence with a non-complementary overlap. The enzyme cleaves at the overlap due to its recognition of the “tail”, and releases the “tail” with a label. This can then be detected. The Invader™ technology is described in U.S. Pat. Nos. 5,846,717; 5,614,402; 5,719,028; 5,541,311; and 5,843,669, all of which are hereby incorporated by reference.
“Rolling circle amplification” is based on extension of a circular probe that has hybridized to a target sequence. A polymerase is added that extends the probe sequence. As the circular probe has no terminus, the polymerase repeatedly extends the circular probe resulting in concatamers of the circular probe. As such, the probe is amplified. Rolling-circle amplification is generally described in Baner et al. (1998) Nuc. Acids Res. 26:5073-5078; Barany, F. (1991) Proc. Natl. Acad. Sci. USA 88:189-193; Lizardi et al. (1998) Nat. Genet. 19:225-232; Zhang et al., Gene 211:277 (1998); and Daubendiek et al., Nature Biotech. 15:273 (1997); all of which are incorporated by reference in their entirety.
“Branched DNA” signal amplification relies on the synthesis of branched nucleic acids, containing a multiplicity of nucleic acid “arms” that function to increase the amount of label that can be put onto one probe. This technology is generally described in U.S. Pat. Nos. 5,681,702, 5,597,909, 5,545,730, 5,594,117, 5,591,584, 5,571,670, 5,580,731, 5,571,670, 5,591,584, 5,624,802, 5,635,352, 5,594,118, 5,359,100, 5,124,246 and 5,681,697, all of which are hereby incorporated by reference.
Similarily, dendrimers of nucleic acids serve to vastly increase the amount of label that can be added to a single molecule, using a similar idea but different compositions. This technology is as described in U.S. Pat. No. 5,175,270 and Nilsen et al., J. Theor. Biol. 187:273 (1997), both of which are incorporated herein by reference.
The ability to perform a variety of preparation and amplification steps in a single miniaturized device has the potential for saving time and expense. Such miniaturized devices can be made much more portable than conventional apparatus, thereby enabling samples to be analyzed outside of the laboratory, such as the location where the samples are collected. A miniaturized DNA analysis device can also allow the analysis steps to be automated more easily. As a result, assays could be performed by less highly trained personnel than presently required.
Thus, there is a significant trend to reduce the size of these sensors, both for sensitivity and to reduce reagent costs. Thus, a number of microfluidic devices have been developed, generally comprising a solid support with microchannels, utilizing a number of different wells, pumps, reaction chambers, and the like. See for example EP 0637996 B1; EP 0637998 B1; WO96/39260; WO97/16835; WO98/13683; WO97/16561; WO97/43629; WO96/39252; WO96/15576; WO96/15450; WO97/37755; and WO97/27324; and U.S. Pat. Nos. 5,304,487; 5,071,531; 5,061,336; 5,747,169; 5,296,375; 5,110,745; 5,587,128; 5,498,392; 5,643,738; 5,750,015; 5,726,026; 5,35,358; 5,126,022; 5,770,029; 5,631,337; 5,569,364; 5,135,627; 5,632,876; 5,593,838; 5,585,069; 5,637,469; 5,486,335; 5,755,942; 5,681,484; and 5,603,351. In addition, there are a number of devices including PCR microchips fabricated on silicon or glass (Wilding et al., 1994, Clin. Chem. 40:1815-18; Shoffer et al., 1996, Nucleic Acids Res. 24:375-79; Cheng et al., 1996, Nucleic Acids Res. 24:380-85; Woodley et al., 1996, Anal. Chem. 68:4081-86; Northrup et al., 1998, Anal. Chem. 70:918-22; Ibrahim et al., 1998, Anal. Chem. 70:2013-17; U.S. Pat. No. 5,498,392 (Wilding et al., 1996), U.S. Pat. No. 5,587,128 (Wilding et al., 1996), U.S. Pat. No. 5,589,136 (Northrup et al., 1996)).
While conventional PCR is performed in volumes of between 10-100 mL and require several hours to process, microchip PCR is performed in volumes of less than 5 mL and can be completed in minutes. The decrease in reaction time for microchip PCR has been achieved as a result of the low thermal mass of silicon reaction chambers and the integration of thin-film heaters (Northrup et al., 1998, Anal. Chem. 70:918).
While silicon microchip arrays have been fabricated for the parallel analysis of multiple samples (Beigrader et al., 1998, Clin. Chem. 44:2191-94), such devices do not facilitate reaction condition optimization. In order to rapidly optimize amplification conditions for a particular target and amplimer pair, an investigator must be able to perform independently controlled, parallel amplifications on a single microchip array. Due to the inefficient well-to-well thermal isolation achievable in arrays constructed of silicon or glass and the complicated fabrication methods required to prepare microchip arrays from such materials, present techniques have not permitted preparation of a cost-effective commercial microchip array for performing such optimization experiments.
Existing apparatus for performing detection reactions such as thermally-controlled biological reactions on a substrate surface are deficient in that they either require unacceptably large volumes of sample fluid to operate properly, cannot accommodate substrates as large as or larger than a conventional microscope slide, cannot independently accommodate a plurality of independent reactions, or cannot accommodate a substrate containing hydrogel-based microarrays. Most existing apparatus also do not allow introduction of fluids in addition to the sample fluid such as wash buffers, fluorescent dyes, etc., into the reaction chamber. Disposable apparatus require disassembly and reapplication of a new apparatus to the substrate surface every time a new fluid must be introduced. Other existing apparatus are difficult to use in a laboratory environment because they cannot be loaded with standard pipet tips and associated pipettor apparatus.
Many existing apparatus also exhibit unacceptable reaction reproducibility, efficiency, and duration. Reaction reproducibility may be adversely affected by bubble formation in the reaction chamber or by the use of biologically incompatible materials for the reaction chamber. Reaction duration and efficiency may be adversely affected by the presence of concentration gradients in the reaction chamber.
Bubbles can form upon introduction of sample fluid to the reaction chamber or by outgassing of the reaction chamber materials. When gas bubbles extend over the substrate surface in an area containing biologically reactive sites, the intended reaction may intermittently fail or yield erroneous results because the intended concentration of the sample fluid mixture has been compromised by the presence of gas bubbles.
Biologically incompatible reaction chamber materials may cause unacceptable reaction reproducibility, by interacting with the sample fluid, thus causing the intended reaction to intermittently fail or yield erroneous results.
Incomplete mixing of the sample fluid can introduce concentration gradients within the sample fluid that adversely impact reaction efficiency and duration. This effect is most pronounced when there is a depletion of target molecules in the local volume surrounding a biologically reactive site. During a biological reaction, the probability that a particular target molecule will bind to a complementary (immobilized) probe molecule is determined by the given concentration of target molecules present within the sample fluid volume, the diffusion rate of the target molecule through the reaction chamber, and the statistics of interaction between the target molecule and the complementary probe molecule. For diagnostic assays, target DNA molecules are often obtained in minute (<picomol) quantities. In practice, it can take tens of hours for a hybridization reaction to be substantially complete at the low target nucleic acid molecule levels available for biological samples. Concentration gradients in the hybridization chamber can further exacerbate this problem.
U.S. Pat. No. 5,948,673 to Cottingham discloses a self-contained multi-chamber reactor for performing both DNA amplification and DNA probe assays in a sealed unit wherein some reactants are provided by coating the walls of the chambers and other reactants are introduced into the chambers prior to starting the reaction in order to eliminate flow into and out of the chamber. No provisions are made for eliminating gas bubbles from the chambers.
There remains a need in the art for methods and apparatus for performing biological reactions on a substrate surface that use a low volume of sample fluid, that accommodate substrates as large as or larger than a conventional microscope slide, that accommodate a plurality of independent reactions, and that accommodate a substrate surface having one or more hydrogel-based microarrays attached thereto. There also remains a need in the art for an apparatus that allows introduction of fluids in addition to sample fluid into each reaction chamber via standard pipet tips and associated pipettor apparatus. There also remains a need in the art for such an apparatus that increases reaction reproducibility, increases reaction efficiency, and reduces reaction duration. There also remains a need in this art for a simple method for removing gas bubbles from such an apparatus. These needs are particularly striking in view of the tremendous interest in biochip technology, the investment and substantial financial rewards generated by research into biochip technology, and the variety of products generated by such research.
Nucleic acid hybridization assays are advantageously performed using probe array technology, which utilizes binding of target single-stranded DNA onto immobilized DNA (usually, oligonucleotide) probes.
The detection limit of a nucleic acid hybridization assay is determined by the sensitivity of the detection device, and also by the amount of target nucleic acid available to be bound to probes, typically oligonucleotide probes, during hybridization.
A common challenge to all DNA hybridization technologies is the lack of control of stringency for each individual probe site. The DNA hybridization process occurs at specific temperature and salinity conditions and varies with DNA sequences. For DNA probe arrays, since the DNA probe sequences are different, hybridization recognition is never perfect under a uniform stringency condition for the entire probe array. The problem is most obvious for short duplexes which often results in single base mismatches. One can minimize the effect of mismatched hybridization by using large probe site redundancy. Stringency control has been provided for each probe site by controlling the electrophoretic movement of oligonucleotides. To successfully implement this later scheme, a meticulously engineered permeation layer is required to prevent DNA molecules or labeling agents being damaged by direct electrolysis or by the product of the electolysis.
In addition, the current DNA array technologies have failed to provide an effective solution to maximize hybridization efficiency. For diagnostic assays, the target DNA molecules are often of minute quantities. The detection limit of the assay is determined by the sensitivity of the detection device, and also by the amount of target oligos bound to the probes during the course of hybridization. In a stationary hybridization chamber where active mixing is absent, the probability of a given target molecule hybridizes to its complementary strand on the surface is determined by diffusion rate and statistics. It takes up to tens of hours for hybridization to complete at low target concentration levels. To better utilize the target molecules and enhance the hybridization, flow through technology has been proposed where the probe arrays are placed perpendicular to the fluidic flow direction. Even with flow through technology, only a portion of the target molecules can come in contact with any specific DNA probe site.
The present invention overcomes the above technical issues by sequentially placing the DNA probe sites in microfluidic channels such that the DNA probe can efficiently contact its binding partner.
U.S. Pat. No. 5,147,607 describes a variety of microassay devices which have microchannels in plastic materials with a reagent such as an antibody or DNA immobilized on the channel at different locations. Techniques for binding antibodies to the microchannel wall are described but techniques for binding DNA are not described. The binding of probes to the microchannel wall does not provide for optimum contact of probe and test sample. U.S. Pat. No. 5,843,767 describes microfabricated flowthrough porous apparatus for discrete detection of binding reactions such as DNA/DNA. WO/98/43739 describes porous flow channels having reagents immobilized in the chamber.
Nucleic acid hybridization chambers are known in the prior art. U.S. Pat. No. 5,100,755 to Smyczek et al. discloses a hybridization chamber. U.S. Pat. No. 5,545,531 to Rava et al. discloses a hybridization plate comprising a multiplicity of oligonucleotide arrays. U.S. Pat. No. 5,360,741 to Hunnell discloses a gas heated hybridization chamber. U.S. Pat. No. 5,922,591 to Anderson et al. discloses a miniaturized hybridization chamber for use with oligonucleotide arrays. U.S. Pat. No. 5,945,334 to Besemer discloses oligonucleotide array packaging.
As currently employed, oligonucleotide array technology does not provide maximum hybridization efficiency. Existing nucleic acid hybridization assay equipment includes numerous components, each of which is a source of inefficiency and inaccuracy.
Hybridization using oligonucleotide arrays must be performed in a volume in which a small amount of target DNA or other nucleic acid can be efficiently annealed to the immobilized probes. For diagnostic assays, target DNA molecules are often obtained in minute (<picomol) quantities. In practice, it can take several (tens of) hours for hybridization to be substantially complete at the low target nucleic acid levels available for biological samples.
In addition, array hybridization is conventionally performed in a stationary hybridization chamber where active mixing is absent. Under these conditions, the probability that a particular target molecule will hybridize to a complementary oligonucleotide probe immobilized on a surface is determined by the concentration of the target, the diffusion rate of the target molecule and the statistics of interaction between the target and the complementary oligonucleotide. Consequently, a larger number of samples must be tested to obtain useful information, and this in turn leads to increased hybridization times and inefficiencies.
In addition, efficiency is increased when the amount of user manipulation is kept to a minimum. As currently performed, oligonucleotide array hybridization requires a great deal of operator attentiveness and manipulation, and the degree of skill required to perform the analysis is high. For example, hybridization is typically performed in an assay chamber, and then data collection and analysis are performed in a separate apparatus (such as a laser scanner or fluorescence microscope). This arrangement requires a substantial amount of handling by the user, and makes the assays both time-consuming and subject to user error.
It is also a limitation of current practice that array hybridizations are performed one array at a time, thereby forgoing the economies of parallel processing and data analysis.
Additional limitations, inefficiencies, and expenses arise from the structural characteristics of existing apparatus. Many existing apparatus are limited in the size of the substrate they can accommodate. Other apparatus are not disposable and therefore require extensive cleaning between runs in order to prevent sample contamination. Yet other apparatus are high mass and therefore not susceptible of the rapid heating and cooling necessary for efficient hybridization. Other apparatus require the use of expensive optics for analysis of the reaction products.
There remains a need in this art for an easy-to-use apparatus for performing biological reactions, particularly nucleic acid hybridization, that comprises a small reaction volume, where the fluid components can be actively mixed, that can be performed in parallel and that minimizes user intervention. There also remains a need for such an apparatus that is easy to manufacture in various sizes, that is disposable to minimize sample contamination, that allows for the use of low cost optical analytical equipment, and that is low mass to allow for rapid heating and cooling of the sample fluid. There also remains a need for methods for using such apparatus to increase hybridization efficiency, particularly relating to biochip arrays as understood in the art. This need is particularly striking, in view of the tremendous interest in biochip technology, the investment and substantial financial rewards generated by research into biochip technology, and the variety of products generated by such research.